KR950009807B1 - Magnetic sensor using emitter injection modulation effect - Google Patents

Abstract

The sensor consists of an emitter region which has the first and second two-sided electrodes supplied by the first bias voltage in horizontal direction, a base region which is supplied by the second bias voltage through the two-sided electrodes of the base electrodes, a collector eletrodes which are positioned symmetrically to the emitter region, and a collector region which produces currents that takes a part in hall effect by the collector electrodes.

Description

Magnetic sensor element using emitter injection modulation effect

1 is a model for explaining the Hall effect.

2A is a cross-sectional structure diagram of a conventional device using a carrier deflection effect.

Figure 2b is a plan view of a conventional device using the injection modulation effect.

3 is a plan view of the device of the present invention.

FIG. 3A is a cross-sectional structural view taken along cut line A-A 'of FIG.

3b is a cross-sectional structural view taken along the line BB ′ of FIG.

4a and b are collector shape diagrams to consider when designing the device of the present invention;

4c and d are explanatory views showing the state of movement of the carrier in the emitter,

5 is a graph showing the relative sensitivity to magnetic fields of the device of the present invention.

The present invention relates to a semiconductor magnetic field sensor, and more particularly, to a magnetic field sensor using injection modulation of a carrier.

Magnetic field sensers have been increasing in demand for information reading of magnetic tapes and disks, shape readings of magnetic silk, and the like for contactless switches and linear or angular displacement detection. In particular, magnetic sensors in various automatic control systems require characteristics such as miniaturization and high reliability. Such magnetic sensors can be roughly classified into the case of using a ferro- material having a permeability greater than 1 and the case of using a material having a permeability of approximately 1 (dia-). In the early field sensors, ferromagnetic materials were used. However, semiconductor manufacturing technology has been used to fabricate magnetic field sensors on integrated chips using the Hall effect inherent in semiconductor devices due to the rapid development of layout or planar technology. It is being pushed in the direction. That is, as shown in FIG. 1, considering the N-type semiconductor device in which current flows in the x direction and the magnetic field B is hung in the z direction, electrons moving at the speed of v are caused by Lorentz force (field Ey). Is deflected in the y direction. In the well known equation qv × B = qEy (where q represents the amount of residual electrons), the Hall voltage Vh can be obtained at both ends of the semiconductor device of FIG. 2 using the electric field value. After all, since the hall voltage is proportional to the magnetic field B, it is used as a sensor for detecting the strength of the magnetic field. The semiconductor magnetic field sensor has already been proved to have a much lower price and superior characteristics than the above-described ferromagnetic magnetic field sensor, and a technology for improving sensitivity to magnetic field has been developed.

The types of semiconductor magnetic field sensors known to date are those using integrated hole plates, field effect transistors (MAGFETs) and bipolar transistors (MT), depending on the structure and principle of operation. : MegnetoTransistor), MD (MagnetoDiode) and CDM (Carrier Domain Magnetometer). These devices have their own characteristics and uses, but in terms of sensitivity, which is the most important factor in magnetic field sensors, MT is known to be the best and its application range is wide. However, since the operation in the MT device is made by a complicated combination between the inherent operation characteristics of the bipolar transistor and the galvanomagnetic effect, the specific mechanism thereof is not fully understood. Due to the ease and excellent properties, the sensitivity has been continuously improved.

The MT element is structurally divided into a vertical MT and a horizontal MT. Vertical MT devices were proposed in 1982 by V.Zieren of the University of Delft, Netherlands (IEE Transanctions on Electron Devices, Vol. ED-29, No.1, Jan, 1982, pp-83-90). Horizontal MT element is a product of IBM's AW in 1982. Suggested by Vinal (see IEEE Transanctions on Electron Devices, Vol. 31, No. 10. Oct. 1984).

Referring to FIG. 2A, it can be seen that the device by Zieren has a bipolar transistor structure having two collectors. Electrons injected from the emitter are deflected to either side by a magnetic field By applied in parallel to the device surface. Since the device of 2a is an NPN bipolar transistor, electrons are deflected toward the collector on the left side, so that the collector current I _C1 on the left side is larger than the collector current I _{C2 on the} right side. That is, the magnetic field strength can be known by detecting the difference between the two collector currents. This uses a mechanism in which the intensity of the magnetic field induces the difference between two collector currents, and is an example in which the operation form of the MT element is explained by carrier deflection.

Meanwhile, referring to FIG. 2B, the collector current values of the left and right sides of the device by Vinal vary depending on which base electrode is activated. Activating the base B1 causes the collector current I _CR on the right to be larger than the collector current I _{CL on the} left, and vice versa when the base B2 is activated. This is because the deflection of the injection current induced by the magnetic field B applied perpendicular to the device surface forms a hall voltage, which causes the bias change in the emitter-base cushion. Thus, the bias change between the actor and the base due to the Hall voltage is detected. This is one example in which the operation form of the MT element is described by emitter injection modulation.

Various MT elements, which are roughly classified as the mechanisms of carrier deflection and injection modulation, as shown in FIGS. 2A and 2B, have been developed to improve sensitivity. For example, in the case of the MT device using the SiMoS technology proposed by RSPopovic in 1986, the relative sensitivity of the magnetic field was realized up to 1.5 / Tesla (which is interpreted as a carrier deflection model). In the case of the SSIMT (Suppressed Sidewall Injection MT) proposed by Ristic, the relative sensitivity to the magnetic field is 30 / Tesla, which represents the highest sensitivity among the devices manufactured so far.

SSIMT is interpreted by the carrier deflection model, and since the two P ⁺ regions on the emitter side are reverse biased with the emitter to narrow the electron injection region, the original collector current is also kept small and the magnetic field The electron deflection by the application of R is relatively large in the injection region, which can realize a greater relative sensitivity than previous devices (IEEE Transanctions on Electron Devices.Vol. 36, No. 6, Jun. 1989, See pp. 1076-1986).

Here, a value, which is generally defined as the relative sensitivity of the magnetic field sensor, that is, the value ΔIc (B) / Ic divided by the collector current divided by the change in the collector current by the collector current, changes linearly with respect to the magnetic field in the carrier deflection model. In the injection modulation model, it changes exponentially with the change of the magnetic field. That is, in the carrier deflection model, in view of the relationship between the difference between the collector currents and the corresponding hall voltages, the hole electric field causing the carrier deflection is determined by the state of the hall voltage by the value of? Ic (B) / Ic. As a result, the sensitivity is obtained as an exponential value because the current density according to the concentration of carriers injected from the emitter is reflected in the hall voltage.

Therefore, in order to achieve high sensitivity in a wide range of magnetic fields, it is desirable to design the device according to the injection modulation model.

However, the conventional devices using the injection modulation model as described above did not realize superior sensitivity as compared to the devices using the carrier reflection model, because the injection current causes the hall voltage to increase and the hall voltage increases. This is because the collector current increased. That is, the conventional injection modulation type magnetic field sensors have not been able to separate the injection current and the injection current, which have superior theoretical advantages but have not been actively used.

Accordingly, an object of the present invention is to provide a sensor capable of separating a current and a line current generating a hall voltage in a semiconductor magnetic field sensor.

Another object of the present invention is to provide a semiconductor magnetic field sensor having a large relative sensitivity to magnetic fields.

In order to achieve the object of the present invention, the magnetic field sensor of the present invention uses the emitter as the active region, which is different from that of the conventional element according to the injection modulation model, using the base as the active region. Two electrodes are used in the emitter to adjust the current to generate the hall voltage, and two base electrodes are used on each side of the emitter so that the bias between the emitter and the base is maintained regardless of the voltage difference in the emitter. Further, in the present invention, the shape of the collector is trapezoidal to suppress locally excessive reverse bias between the base and the collector due to the voltage difference in the base.

Hereinafter, a preferred embodiment of a semiconductor magnetic field sensor of the present invention will be described in detail with reference to the accompanying drawings.

Referring to FIG. 3, the magnetic field sensor according to the present invention includes an emitter E having two electrodes UEE and DEE, and collector electrodes LCE and RCE positioned at both sides of the emitter. Four base electrodes (ULB, URB, DLB, DRB) located on the left and right collectors LC and RC on the left and right sides and respectively provided on the left and right sides of the emitter, respectively; And a semiconductor substrate (SUB) as a base formed by integrating them. In particular, the shape of the collector is characterized in that the width is narrowed toward the lower base (DLB, DRB).

Also note that the length (L) and width (W) of the emitter are important factors in the design of the device. The shape of the collector and the size of the emitter will be described in detail below. The emitter supply voltage V (1at) is applied to the upper emitter electrode UEE, and the lower emitter electrode DEE is connected to the ground. If the semiconductor substrate SUB is a P-type silicon substrate having an orientation of (100), the emitter E and the chelators LC and RC become N-type wells, and the emitter electrodes UEE and DEE ) And collector electrodes LCE and RCE are shallow diffusion regions of N ⁺ type, and the base electrodes ULB, URB, DLB, DRB are diffusion regions of P ⁺ type. The size of the emitter, collector and base, or the distance and spacing between these elements, should be configured to achieve maximum sensitivity.

For example, the size of the device is 76μM in width and 48μM in length, the emitter is 18x24, the emitter electrode is 18x12, the base is 20x12, the collector is 7, the narrow width is 5, and the length is 20. The collector electrode is 13x20. In addition, the distance between the emitter and the collector may be 9, and the distance between the collector and the base may be 6.

3A and 3B are views showing cross-sectional structures different from incision lines A-A 'and B-B' of FIG. FIG. 3A shows the cross-sectional structure of the emitter, the collector and the collector electrode, and FIG. 3B shows the cross-section of the emitter, the emitter electrode and the base. The N wells forming the emitter and the collector have a depth of about 4 micrometers, and the emitter electrode, the collector electrode, and the base have a shallow section of about 2 microns.

The design of the length (L) and width (W) of the emitter should be made with careful consideration of how the Hall voltage varies with the geometrical factors of the device. The geometric factor is a value representing the influence of the input and output electrodes in the device and is generally expressed as the W / L ratio of the emitter. It can be seen that the emitter used in the present invention is W / L = 18/24 = 0.75. However, this value may be set differently under conditions that may cause the maximum hall voltage. The hall voltage V _H represented by the output voltage from the collector is

V _H = μ _n × (W / L) × V _1at × G × B... … … … … … … … … … … … … … Formula (1)

Expressed as Where _n is the mobility of electrons, G is about 0.8, and B is the magnetic field applied perpendicular to the surface in FIG.

On the other hand, the relative sensitivity Sr (relative sensitivity) is the current density variation ΔJc = Jc (B) -Jc (in the collector before [J _c (0)] and after [J _c (B)] before the magnetic field B is applied. 0) and the current amplification gain β of the bipolar transistor. That is, when Sr = [ΔJc / Jc (0)] × β ⁻¹ , and the relation of Jc (0) and Jc (B) obtained by a well-known transistor equation is pressed against the equation of Sr, Sensitivity Sr is finally

Sr = So x [EXP ( _q V _H / kT) -1] x β ^-1 . … … … … … … … … … … … Formula (2)

Expressed as Here, So is a constant term representing absolute sensitivity and can be approximated by the equation (2). As a result, it can be seen from Equation (2) that the relative sensitivity Sr is a value having an exponential proportionality according to the hall voltage V _H. This was mentioned earlier. Referring to Equation (1), it can be seen that the Hall voltage V _H is changed by the voltage V _1at applied to the upper electrode UEE of the emitter in a state where the emitter W / L ratio is determined. To determine the value of V _1at , consideration should be given to the influence of the electric field E _1at formed in the longitudinal direction of the emitter. Here, since the magnitude of the electric field is a major factor influencing the injection of the carrier in the transistor, it should be noted that the preferred setting thereof is an important sign in achieving the object of the present invention using the injection modulation model. The magnitude of the electric field E _1at is not represented by a change in the relative sensitivity Sr but by a change in the geometrical factor of the emitter under the same V _1at . For example, if V _1at is 10V and the emitter length L is 10 μM and 1 μM, respectively (assuming the emitter width W is constant), the magnitude of the field E _1at is 10 ⁴ V / M and 10 ⁵ V / M, respectively. .

4A and 4B, the equilibrium drifting and injection states of electrons in the emitter can be known according to the magnitude of the electric field E _1at formed in the longitudinal direction of the emitter. In the drawing, _1at GE represents the slope (gradient) in the vertical direction of the dislocation by GE _1at, and Gφ _EB shows the tilt in the horizontal direction by the emitter to the potential barrier (potenial barrier) between the base. The slope Gφ _EB between the emitter and the base may be regarded as about 0.1 eV, which is the potential of the PN junction without change in the case of the forward bias. The potential size between emitter and base affects the state of equilibrium drift. On the other hand, the GE _1at depends on the size of the electric field, as shown in Figure 4a, in the small probe so that the equilibrium drifting and injection of the electron is generated, but as shown in Figure 4b the electron energy will overcome the potential between the emitter and the base When the electric field is large enough, the motion of the electrons is almost injecting. In this way, the range of the electric field E _1at applicable to the device of the present invention can be set on the basis of the magnitude of the electric field E _1at and thus the mobility of the electrons. That is, it is limited in the range that the equilibrium drifting of electrons and the emitter-base injection occur simultaneously (the current due to the equilibrium drifting of electrons is added to maximize the hall voltage without increasing the injection current any more). this is,

Mean free stroke of the former × E _1at <0.1 eV ×. … … … … … … … … … Formula (3)

It can be represented as. In the floating equation of Formula (3), "left side" is the energy obtained by the electron by the electric field E _1at , and r is a quantum mechanically determined constant, which is preferably about 2-3. As a result, when the energy of electrons by E _1at ("left side" of Equation (4)) is close to 2-3 times the potential between the emitter and the base, the maximum of V _1at is applied and the hall voltage V _H is maximized [ Eq. (1)], whereby the maximum value of relative sensitivity SR can be obtained (see Eq. (2)). If V _1at is applied _at 10V, the value of the “right side” of the above formula (3) and the mean free path of the electrons can be known, so that the range of E _1at can be known. This makes it possible to set the appropriate length L of the emitter. If the ratio of W / L is 0.75-0.77, the geometric value of the emitter can be determined by obtaining the value of the emitter width W using the obtained L value.

On the other hand, the decision on the shape of the collector should start with consideration of the effective base width W _B according to the bias between the base and the collector and the punchthrough according to the reverse bias in the collector-base cushion. In FIG. 3, bias voltages of 12V and OV are applied between the upper electrode UEE and the lower electrode DEE of the emitter, respectively, and the collectors of 12.7V and 0.7V are respectively applied to the upper and lower bases UBR (DRB). It is assumed that a bias of 30V is applied to (RC). Between the emitter and base, the PN section is forward biased by a threshold voltage of 0.7V. The base-collector is 17.3V (= V _CBU ) for the upper section (URB-RC) and 29.3V (V = _CBD ) for the lower section (DRB-RC), and is reverse biased between the base and the collector. Thus, as shown in FIG. 4C, the depletion region is formed wider in the cushion (V _CBU &lt; V _CBD ) having a larger reverse bias. This causes each effective base width W _B at the two bases to be different. When the effective base width at each base is different, the amount of injection current of electrons at the emitter is changed, which acts as a factor for changing the collector current, which is another way in which the present invention analyzes the desired device characteristics. It becomes an obstacle.

Therefore, in order to make the effective base width constant, as shown in FIG. 4D, the collector width at the lower section DRB-RC is made smaller by ΔX than the upper side. By reducing the width of the collector by the width of the depletion area due to the difference in cushion bias, the effective base width of the bases can be kept the same. In addition, since the reverse bias voltage at the lower collector section is relatively high, a punch-through occurs between the collector and the base at the lower section section, thereby increasing the probability of shorting. Therefore, the distance between collector-base d2 in the lower collector section must be larger than the distance between collector-base d1 in the upper collector section. Those skilled in the art that the width of the collector, the collector-base distance, and the emitter-base distance can be appropriately set by the above-described processes for determining desirable characteristics for the device of the present invention. If you know it well.

Then, as described above, look at the operation of the magnetic field sensor configured for the geometrical factors for the device of FIG. When V _1at applied to the emitter is set to 12V, the most Bz is applied to the device in the vertical direction (z direction), the electrons in the N-type emitter are deflected to the right by the Hall effect. Then, a hole electric field is generated from the left side of the emitter toward the right, and the hole electric field is large enough to cancel the flow of the hole current induced by the deflected electrons. If a voltmeter is placed between the left and right sides of the emitter, the hall voltage V _H will be measured by multiplying the hol field by the emitter width W. Since the emitter-base is forward biased, the excess carriers in the emitter energized by the field E _1at are injected into the base over the potential barrier (lowered by the forward bias) formed at the PN junction between the emitter and the base. do. The current generated by the injection of electrons from the emitter, that is, the injection current, is constant because the voltage V _EB between the emitter-base is held constant and the effective base width is constant. However, the Hall voltage V _H formed in the emitter by the applied magnetic field is added to the emitter-base voltage and reflected in the collector current density variation ΔJc described above, and as a result, the relative sensitivity Sr is expressed by the above-described formula ( According to 2), it is exponentially increased by the hall voltage V _H. How much of the Hall voltage can be induced for a given magnetic field has been described above.

Referring to FIG. 5, it can be seen how the relative sensitivity Sr is represented according to the magnitude of the voltage V _1at applied to both ends of the emitter in the magnetic field sensor device of the present invention as shown in Equation (2), according to Equation (2). Can be. For example, a relative sensitivity of about 100 / T appears when a magnetic field of about 0.55T is applied _at V _1at of 12V. If a magnetic field of 1.55T or more is applied at 12V, very high state sensitivity can be obtained by the exponential characteristic of the curve.

In order to fabricate the magnetic field sensor device of the present invention as shown in FIG. That is, first, in order to form N wells (which become emitter and collector regions) on the substrate, ion implantation and diffusion are carried out at a dose of 10 ¹⁷ / cm 2 using a first port resist pattern on a P-type silicon substrate. The second and third photoresist patterns are used to form shallow P ⁺ and N ⁺ regions (which are emitter electrodes, collector electrodes, and bases), and to form contacts and electrodes and metal wiring patterns with the electrodes. Fourth and fifth photoresist patterns are used. In such a process, the geometrical application of the above-described devices must be taken into consideration when designing the mask.

As described above, the present invention has the effect of providing a magnetic field sensor element that can preferably utilize the injection modulation model that can obtain a greater sensitivity than the carrier deflection in the semiconductor magnetic field.

In addition, the present invention has the advantage of providing a magnetic field sensor element that can achieve a high sensitivity compared to the conventional in a simple process.

Claims (3)

A semiconductor magnetic field sensor element responsive to an applied magnetic field, comprising: a second conductive element formed on a substrate of a first conductivity type and having first and second opposite electrodes provided with a first bias voltage in a horizontal direction with respect to the substrate; And an emitter region of the type: first, second, third and fourth base electrodes symmetrically arranged on the first, second, and second electrodes of the emitter region, respectively, through the bielectrodes of the base electrodes. A base region as the substrate to receive a second bias voltage different from the first bias voltage, and a collector electrode positioned between the bielectrodes of the base electrodes and symmetrically disposed in the emitter region, respectively; And having a collector region of a second conductivity type which generates a current due to a Hall effect through the collector electrodes when a magnetic field is applied in a direction perpendicular to the substrate. Semiconductor magnetic field sensor element, characterized in that the voltage responsive to the magnetic field of the size to be formed on the emitter region. The semiconductor magnetic field sensor device of claim 1, wherein the second bias voltage has a voltage level higher than that of the first bias voltage. 3. The semiconductor magnetic field sensor element according to claim 2, wherein the collector region has a trapezoidal width.